Search Swinburne Research Bank
Please use this identifier to cite or link to this item: http://hdl.handle.net/1959.3/44081
 Download PDF (Published version) (Adobe Acrobat PDF, -1 bytes) |
- Title
- Neuron splitting for efficient feature map formation
- Author(s)
- Andrew, Lachlan L. H.
- Abstract
- Kohonen's Self Organizing Feature Map (SOFM) produces an ordered mapping from one space to another. This paper describes an algorithm inspired by the splitting initialization for the classical LBG method for vector quantizer design, which allows the efficient generation of maps with various topologies and with high local and global ordering.
- Publication type
- Conference paper
- Source
- Proceedings of the 2nd Australian and New Zealand Conference on Intelligent Information Systems, Brisbane, Queensland, Australia, 29 November-02 December 1994, pp. 10-13
- Publication year
- 1994
- Keyword(s)
- Algorithms; Classical LEG method; Computational complexity; Feature map formation; Global ordering; Kohonen Self Organising Feature Map; Local ordering; Mathematical models; Neural networks; Neuron splitting; Ordered mapping; Probability density function; Self organising feature map; SOFM; Splitting initialisation; Unsupervised learning; Vector quantisation; Vector quantiser design; Vectors
- Publisher
- IEEE
- ISBN
- 9780780324046, 0780324048
- Publisher URL
- http://dx.doi.org/10.1109/ANZIIS.1994.396960
- Copyright
- Copyright © 1994 IEEE. Paper reproduced here in accordance with the copyright policy of the publisher.
- Additional information
- While performing this work, the author was on a scholarship from the Australian Telecommunications and Electronics Research Board (ATERB).
- Full text
- Peer reviewed
